Description
A Concept and Prototype for Linking Business Intelligence To Business Strategy
PUZZLE: a concept and prototype for linking
business intelligence to business strategy
Kamel Rouibah
*
, Samia Ould-ali
Department of Quantitative Methods and Information Systems, College of Business Administration,
P.O. Box 5486 Safat 13055, Kuwait
Received 29 November 1999; revised paper accepted 18 March 2002
Abstract
Business intelligence (BI) is a strategic approach for systematically targeting, tracking,
communicating and transforming relevant weak signs
1
into actionable information on which stra-
tegic decision-making is based. Despite the increasing importance of BI, there is little underlying
theoretical work, which directly can guide the interpretation of ambiguous weak signs. This paper
gives an insight into the issue through a new strategic business intelligence system called PUZZLE.
We describe this system and validate it by designing a prototype, test the system using in-depth
interviews, and hold learning sessions in order to further knowledge about BI. The main results from
tests show that: interpreting weak signs is potentially important for senior managers, consultants, and
researchers; interpretation can be achieved gradually by bringing the weak signs together using a
tracking form based upon the concept of actor/theme/weak signs/enrichment /links; interpreting
weak signs is a complex process of establishing links between the weak signs. Final results show
that the individual cognitive process appears heuristic when interpreting weak signs. Implications for
strategic management practice and research are addressed. q 2002 Elsevier Science B.V. All rights
reserved.
Keywords: Business intelligence; Weak signs; Interpreting weak signs; Fast response management; Strategic
information system; Strategic business intelligence system; Creativity; Ill-structured problem; Research engi-
neering; Exploratory research; Prototyping
Journal of Strategic Information Systems 11 (2002) 133±152
0963-8687/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved.
PII: S0963-8687(02)00005-7
www.elsevier.com/locate/jsis
* Corresponding author. Tel.: 1965-484-2671x8445; fax: 1965-254-9408.
E-mail address: [email protected] (K. Rouibah).
1
The authors use the concept of weak sign instead of `weak signal' as proposed by Ansoff, 1975 because they
feel the word `signal' implies greater quantitative measurability. However, authors continue to use it for the same
purpose.
1. From business intelligence to interpreting weak signs
Companies are evolving in turbulent and equivocal environments (Drucker, 1993;
Kelly, 1998; Grove, 1999). This requires companies to be alert and watchful for the
detection of weak signals (Ansoff, 1975) and discontinuities about emerging threats and
opportunities and to initiate further probing based on such detection (Walls and
Widmeyer, 1992. In such environments, business intelligence (BI) is surfacing to deal
with the large volume of information available but which are often misleading, inaccurate
and untimely (Martinsons, 1994; Futures Group, 1997; Attaway, 1998; Herring, 1998;
Freeman, 1999; Groom and David, 2001). The main crucial question raised by companies
in such environments is how to exploit these information elements to grasp opportunities,
and avoid surprises when discontinuities occur (Grove, 1999; Moore and McKenna,
1999). This is the reason that companies need to have a well analysed, designed, and
developed strategic business intelligence system (SBIS) (Martinsons, 1994). The emphasis
here is on information systems that enhance strategic decision-making and that support the
competitive strategy of an organisation (Wiseman, 1988). Much has been written on
environmental scanning systems since Aguilar, Ansoff and Porter (e.g. Beal, 2000).
However, it seems that the growing uncertainty of business environments still raises the
need for an ef®cient SBIS to support scanning and interpreting information so that
valuable intelligence may be delivered to senior managers. This paper develops a new
SBIS and its implementation as a computer system. It is oriented to improve the BI and to
ensure success of the business strategy. The meaning of BI emphasised in this paper,
termed in French as veille strateÂgique, is considered as a systematic approach by which
a company keeps itself vigilant and aware of developments and early warning signs in its
external environment in order to anticipate business opportunities or threats. The external
environment includes all factors and events outside the company that can affect its
performance. Designing a successful SBIS requires an understanding of the relationship
between the BI process and weak signs. There are several similar variants of BI processes
(Martinsons, 1994; Attaway, 1998; Nolan, 1999). Our intent is not to describe in detail
those approaches but only to show the link to weak signs. The description of these similar
processes is beyond the scope of this paper. Among these, one is certi®ed ISO 9001 and is
mainly oriented toward weak signs management. This process is explained below.
1.1. Focus of this research
We consider the BI process as cyclical, involving ®ve phases (Fig. 1, adapted from
Lesca, 1994).
The ®rst phase `targeting' consists of bounding the surveillance of the company's
environment to set tracking priorities. The second phase consists of organising tracking
and selecting the crucial weak signs. The third phase consists of routing the weak signs
collected from outside to inside the organisation. The fourth phase `interpreting' consists
of transforming the collected information into actionable intelligence. If interpretation is
signi®cant, actions can be taken in phase 5. Otherwise, information search has to be re®ned
in a more speci®c way (return to phase 2) if information is imprecise or; (2) the boundary
(target) has to be rede®ned (return to phase 1) if it is too large. Of these ®ve phases, the
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 134
fourthÐinterpretingÐis the most important and most dif®cult (Rouibah, 1998; Attaway,
1998; Subramanian and IsHak, 1998). Inadequate transformation leads to failures in using
BI (Martinsons, 1994; Lesca and Caron, 1995).
Much has been written about the characteristics associated with events as threats or
opportunities, their consequences and people's reactions (Thomas and McDaniel, 1990;
Schneider and Meyer, 1991; Wagner and Gooding, 1997). However, researchers have not
directly examined how people interpret weak signs in order to identify whether they turn
into opportunities or threats even though interpreting weak signs is recognised as an
important research topic for strategic decision-making (El Sawy and Pauchant, 1988;
Thomas and McDaniel, 1990; Ansoff and McDonnell, 1990; Martinsons, 1994; Freeman,
1999). In addition, despite the need for strategic intelligence delivery being widely
recognised (Attaway, 1998; Herring, 1998; Nolan, 1999; Freeman, 1999; Groom and
David, 2001), the problem of interpreting weak signs has not received much attention
in the academic world.
While providing insight into the problem was an operational objective of this study,
there were also important theoretical and practical questions that had to be asked:
1. How to interpret weak signs in order to create meaningful maps of environmental
changes useful to managers?
2. In what format should weak signs be presented?
3. How can such maps be created?
From a practical perspective, answers were sought to discover what dif®culties and
operational questions are associated with the interpretation of weak signs.
Hopefully, this paper provides some insight into those issues. The remaining segments
of the paper are organised as follows. In the next segment dif®culties of interpreting weak
signs and research methodology are discussed. In the third segment, relevant and current
literature on the subject of this paper is reviewed. In the fourth segment of the paper, a new
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 135
Fig. 1. Phases of business intelligence.
prescriptive approach to deal with weak sign and the speci®cation of the computer tool's
functionality will be introduced. The ®fth segment presents prototype capabilities and
limitation. The sixth segment of the paper deals with lessons learned from validation of
the approach. The concluding segment of the paper offers the implications of our ®ndings
for both academics and practitioners.
2. Work context and research methodology
Before describing the research methodology used, Sections 2.1±2.3 discuss the context
of weak signs and the concept of strategic information systems (SIS).
Two reasons explain the dif®culty in interpreting weak signs: the position of such
information in strategic decision-making, and the nature of the information involved.
2.1. The position of weak signs in decision making
Two generic modes of scanning can be acknowledged: reactive and proactive. In the
reactive mode of scanning (Cyert and March, 1963) search is stimulated by a problem and
directed towards supporting strategic decisions. Its bene®ts are readily apparent to
managers because the information collected could be directly integrated into strategic
decision-making. However, the proactive mode of scanning or surveillance is exploratory
and not directed to any prede®ned problem (Aguilar, 1967). In the latter case, the required
information cannot be de®ned in advance. Its bene®ts are less easily measured. Therefore,
monitoring activities are not fully integrated into the decision-making process (Lesca and
Caron, 1995), senior managers continue to make poor decisions (Martinsons, 1994), and
weak signs are ignored (Rouibah, 1998). Weak signs associated with the proactive mode
may lead to a change in business strategy and therefore will require transformation into
actionable intelligence before the information becomes usable. The proactive mode is the
main concern of this paper.
2.2. The nature of information involved
The particular characteristics of weak signs make them very dif®cult to manage.
According to El Sawy (1985) and Rouibah et al. (1997), a weak sign is:
² anticipatory: it informs about changes when they begin to occur,
² uncertain: it concerns events that have not yet been realised,
² ambiguous: its content is usually uncertain or could be deliberately contaminated or
distorted (for example by a competitor),
² fragmentary: each information element taken alone is insigni®cant; however,
signi®cance increases gradually when combined with other weak signs,
² dynamic: it evolves over time,
² cyclical: it has a complex life cycle from growing to declining that varies in its duration
and signi®cance (Ansoff, 1975; McCann and Gomez-Meija, 1992),
² qualitative: in most cases, numbers are not involved and information may be available
in other forms such as written, verbal, or visual images.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 136
Weak signs are thus subject to perception and interpretation, and multiple meanings are
possible. In order to avoid any confusion in the use of the word signal, we prefer to call
such information a weak sign. Here, we de®ne weak signs as uncertain and fragmented
information about developments and trends. These have not been completely realised, or
they have potential consequences, or are perceived to have a signi®cant impact on
organisational performance, either as threats or opportunities.
Taking into account the previous explanation, it is clear that quantitative and archival
information available in on-line computerised databases (OCDBs) (McCann and
Gomez-Meija, 1992) are out of the scope of this paper. They can help to track changes
in world markets and to decrease the time needed to discover trends. OCDBs rely on
information that can be codi®ed and entered textually or numerically. They are used in
combination with well-known scientometrics and bibliometric techniques (e.g. see
Courtial, 1994). These are more concerned with how often archived information is
consulted, and are very often used to perform trend analysis in research (e.g. assessment
of laboratory results, comparison between laboratories). Nevertheless, these tools are
inappropriate to assess unwritten information, which come from personal sources as
some weak signs may.
Based on the previous dif®culties, it can be assumed that a manager's attitude towards
BI would be more effective if a SBIS were developed to assist with a better handling of
weak signs. Section 2.3 explains what we mean by a SBIS.
2.3. Strategic information systems for business intelligence (SBIS)
Strategic information systems (SIS) are systems which contribute signi®cantly to an
organisation's performance and, consequently, to play a major role in its competitive
strategy. A SIS is de®ned as a system having a profound effect on business success by
in¯uencing or shaping a company's strategy or by playing a direct role in the implementa-
tion of that strategy (Sabherwal and King, 1995; Galliers, 1991). An alternative interpreta-
tion of SIS suggests that it is not necessarily a particular information system, but rather a
combination of those parts of an organisation's cluster of information systems, which feed
its strategy planning processes (Clarke, 1994). Among factors that in¯uence the SIS
application, external factors (such as environmental competition, uncertainty and external
needs) are driving forces for any SIS application (Choe et al., 1998). A SIS-oriented
toward external changes helps an organisation to remain competitive and proactive.
As a result, all components of information systems, which support all or part of BI
phases (Fig. 1) are considered to be SIS. If the above is agreed, then we consider a SBIS as
SIS oriented to support company's BI activities.
2.4. Research methodology
Interpreting weak signs is a real problem encountered by many managers in different
organisations that is closely related to the strategic decision-making process. This process
is recognised as ill-structured and not adequately understood (Walls and Widmeyer, 1992;
Schneider and Meyer, 1991; Channel et al., 1997). While the current situation is known
(dif®culty of the interpreting weak signs), the target or ideal situation that informs
managers about potential threats or opportunities is unknown. Surveys (collecting large
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 137
amounts of data followed by testing hypotheses) are not appropriate for helping managers
in this situation. Therefore, the chosen methodology used to perform this research has been
called `engineering research' (Channel et al., 1997). It is an on-site methodology useful
for expanding knowledge about complex and ill-structured processes, such as interpreting
weak signs, and helping managers to progress.
Characteristics of this research methodology are as follows:
1. Engineering activities: the researcher is not an observer, but is a part of the problem
solving process. He acts as an engineer who develops a conceptual model, builds a
prototype, acts as a facilitator and evaluates his construction on site in the organisation.
2. Exploratory research: due to the lack of structure underlying the problem addressed, it
can only be approached by exploratory research based on trial and error, in modelling,
building, testing and evaluating.
3. Creation of procedural knowledge: it refers to the knowledge that helps to act and to
reason (the knowledge of what to do next). This takes the form of a method, a guideline
or a prototype.
The main results expected from this methodology fall into the following:
1. Ability to stimulate interest. Most senior managers or executives needing to be
interviewed are not always able to express clearly either their dif®culties or their
requirements. Therefore, it is useful to support them with `something' in the form of
a prototype in order to attract them and motivate their meeting with researchers.
2. Prototype. This can be used as a support for training and learning of users who will
learn faster by doing rather than by talking. We improve the prototype using users'
suggestions.
3. Creation of new information. Testing the prototype enables the collection of users'
reactions and relevant observations, which would be out of reach without the prototype
or when traditional methods such as surveys are used.
4. Open new research directions. Negative users' reactions during tests become new
research directions and new concepts to develop. The feedback (data collected during
tests with the end users) can be re-used to improve the interpreting process in order to
support managers' needs and contribute to more theoretical knowledge in the ®eld of
BI.
5. Re®nement of concepts. Based upon users' reactions and suggestions, we go back to
the concepts (artefact) we developed in order to further knowledge and propose
enhancement and re®nement.
Section 3 speci®es the appropriate literature with regard to advantages and limitations to
the model developed.
3. Literature review
Several researchers admit that the information collected must be transformed into
actionable intelligence before it becomes usable (Martinsons, 1994; Lesca and Caron,
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 138
1995; Herring, 1998; Koneig, 1996; Attaway, 1998). Moreover, to achieve such an
objective, authors propose to integrate unrelated pieces of information in order to create
a holistic picture (Valette, 1993; Martinsons, 1994; Lesca and Caron, 1995; Koneig, 1996;
Attaway, 1998; Subramanian and IsHak, 1998; Freeman, 1999).
However this transformation has received little attention. Only two methods 3T (El
Sawy and Pauchant, 1988) and PUZZLE (Valette, 1993), to our knowledge, have
addressed the problem. According to those authors, interpreting weak signs can be done
based on a tracking form composed of `actor/theme
2
/information'. Pieces of information,
related to a theme and a speci®c actor (e.g. a competitor), can be pieced together to get a
mental map `puzzle'. The term `puzzle' has been used by several researchers to create an
analogy between interpreting weak signs and reconstruction of a puzzle (see Gilad and
Gilad, 1986; Valette, 1993; Lesca and Caron, 1995; Attaway, 1998; Hall, 2001). Valette
suggests using PUZZLE to detect signi®cant environmental changes and to orient the
scanning. El Sawy and Pauchant found collective interpretation produces more value
than when individuals work alone. They also found that a shift in the interpretation is
generated through the perception of new weak signs or the occurrence of new learning and
ideas. Moreover, a change in a cognitive map takes place through cognitive functions,
which are invariant (content independent), and thus, can be applied to a variety of tasks
and situations. However, these cognitive functions have not been studied and implemented
as a guideline. We propose to reuse the idea of integrating bits of information together in
the form of puzzles and to build a new methodÐaccompanied by a prototypeÐin line
with the two previous ones, which we continue to refer to as PUZZLE. In order to
complete the understanding of weak signs and the cognitive functions that occur during
the interpretation we turn to creativity and learning.
According to several authors output and process are key concepts in creativity (e.g.
Couger et al., 1993; MacCrimmon and Wagner, 1994; Bostrom and Nagasundaram, 1998).
The output is built up gradually by successive alteration (Plsek, 1996). One process that is
common to many theoretical contributions is making connections or associations (see
Stenberg, 1988; Kanter, 1988; MacCrimmon and Wagner, 1994). This process is one of
the main focus of this paper; it refers to the creation of new ideas through relationships
between existing ideas. In our model, such connections can be triggered by linking pieces
of information together in order to infer new information from adjacent ones; and to
generate hypotheses that must be validated quickly. This idea is derived from the
kaleidoscopic thinking of Kanter (1988). It allows the creation of arrangements by
connecting fragmentary information elements until new patterns and actions can be
generated. Such idea is also in line with Herring's view of intelligence as a process in
which information is subject to systematic examination and determination of signi®cant
relationships (Herring, 1998). Weak signs are not all alike since they probably come from
different sources. Thus, they can support or contradict each other, or one weak sign may be
the cause, the consequence, or the result of another weak sign. The process of making
connections was implemented by connecting weak signs through a typology of links:
causality, con®rmation, and contradiction.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 139
2
The term `theme' is used in cognitive psychology and stresses a dynamic description rather than a static point
of view (El Sawy and Pauchant, 1988).
The limited rationality theory (Simon, 1960) gives insight on the human interpretation
of issues. According to this theory, the human brain has limited capacities for interpreting
information and it is in¯uenced by the nature of the information. Miller (1956) showed that
the maximum number of chunks (weak signs in our model) that could be remembered and
dealt with at one time in short-term memory, is 7 ^2. Based on Miller's ®ndings, Simon
proposed a reasoning mode that is heuristic. This requires a manager to build partial
environmental maps. These help him to act even though they are incomplete. In addition,
according to research in cognitive theory, individuals treat information better and faster
when it is presented visually (Meyer, 1991). Therefore, we have chosen to represent the
interpreted information as a graphical map, that we continue to refer to as puzzle. Puzzles
are composed of weak signs and their links. During the creation of a cognitive puzzle,
different modi®cations can re¯ect what has been learned. Piaget (1996) and Norman
(1982) identi®ed the following modes of learning: assimilation where new information
is assimilated into an old cognitive map; accommodation where old cognitive maps are
modi®ed when new information does not ®t the old cognitive map; and structuring when
new cognitive maps are formed. Then, it can be inferred that elements of a puzzle will be
subject to addition, deletion, and mergers.
Based on that previous observations it should now be clear that there is a very intimate
and dynamic relationship between interpretation of weak signs and creativity.
The following method was developed from the previous observations.
4. Proposed method
We de®ne the interpretation of weak signs as both a process and product:
² a process by which an individual or group of individuals transform weak signs into
meaningful maps, even partial, about how a company's present and future business
environment may be impacted,
² an end product, that is the output of that process.
Fig. 2 shows the conceptual model for the interpretation.
4.1. Phases of the method
We propose a method based upon a creative process of seven steps.
4.1.1. Step 1Ðinput to the method
One way to establish boundaries of the environment is to limit the surveillance effort to
(a) the speci®c targets (actors
3
/theme) and; (b) to key strategic information sources. Weak
signs collected according to targets are input to the method. These information elements
are in their early stage and have been collected according to surveillance mode.
Interpreters are non-experts.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 140
3
Actors are all factors and events outside the company that affect its performance, e.g. suppliers, competitors.
4.1.2. Step 2Ðenrichment of the weak signs
All employees of a company could be a potential source of valuable intelligence. A
single information element may lead to signi®cantly different interpretations based on its
context. One might enrich an element in different ways: assess its content using certain
criteria (surprise, validity, relevance, timeliness, completeness, accuracy), provide
continuations or contradictions to the existing information, and to complete missing
information. The value of a weak sign is enhanced when all the views are integrated.
4.1.3. Step 3Ðcategorisation of weak signs
Once the content of a weak sign has been evaluated, it can be categorised and classi®ed
according to themes. These can be identi®ed during a learning session and bounded by an
initial de®nition. A theme is a subject of interest that informs about an external actor for
example the R&D policy of a competitor.
4.1.4. Step 4Ðcreation of puzzles
This step consists of selecting certain information elements from the themes in order to
elaborate puzzles. The construction of a puzzle is a real act of creativity because it consists
of linking weak signs together in such a manner that fragmentary elements become an
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 141
Fig. 2. Framework of weak signs interpretation.
intelligible map, even if incomplete. A puzzle is a visual map, focused on a speci®c actor,
where its nodes are small sentences (phrases) corresponding to weak signs; and edges are
reasoning links (con®rmation, contradiction, causality) which connect different nodes.
Puzzles are useful to perceive a change when they are compared at different times.
4.1.5. Step 5Ðperception of modi®cation
Different transformations may occur when examining a initial puzzle called G
0
:
1. delete information in G
0
, if it is obsolete;
2. subdivide rich information of G
0
;
3. merge two information elements of G
0
into one element when they are identical or one
of them is redundant;
4. add a link between two information elements of G
0
i.e. one has been identi®ed as a new
link such as a continuation;
5. delete a link between two information elements of G
0
4.1.6. Step 6Ðre®nement of puzzles
Each puzzle must be ¯exible, and continually restructured as new weak signs are
collected. Examination of a puzzle based upon new information could challenge the
existing information in contradiction, con®rmation, or continuation. Thus, other
supplementary transformations can take place in the puzzle G
0
, other than the ®ve
above: modi®cation of an information content; substitution of information (new
information replaces existing).
4.1.7. Step 7Ðproposal of actions
Going from atomised pieces of information to a re®ned puzzle helps users to reason.
Arguments and questions are required to conduct such reasoning process:
² Infer new information from the related and adjacent elements that needs immediate
validation.
² Provide progressive veri®cation of the coherence between information elements.
² Orient the scanning and tracking of new and missing information.
4.2. Functions needed for implementation
According to our model, creation of puzzles requires an appropriate computer tool that
is composed of a database and an ideas mapping tool to support cognitive mapping. While
the database will store, and retrieve information, and capture users' knowledge during
enrichment activities (pull for enrichment and push for discussion); the drawing graph will
support the creation and analysis of puzzles. Co-ordination and communication are also
necessary functions for collective interpretation.
A form will record the following for each information element:
² a title, this is a short anticipatory information about actions of an actor,
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 142
² a tracker that collects the information,
² a theme, to which information elements are related,
² an information source that generates information,
² each information element is enriched,
² each information element is evaluated in terms of importance/reliability so that differ-
ent representations can be generated according to their importance and reliability.
² each information has an editing date,an expired date and a date for the expected event,
² an end user in the organisation who needs the information.
To ensure the security and con®dentiality of the system, each user should have
identi®cation and privileges in storage, enrichment, extraction of information, and access
to a puzzle summary.
5. Prototype capabilities and limitations
The prototype developed, called PUZZLE, comprises Lotus Notes (LN) and Decision
Explorer (DE) (from Banxia). DE was chosen because it can be easily customised to
generate puzzle maps. On the other hand, LN was chosen for three main reasons.
² LN, as a technology platform, is able to create many applications. It contains various
sophisticated tools and an integrated macro language, which allows further
development and customisation of the applications.
² LN, as groupware, allows several users to work simultaneously on the same application.
Through its shared databases and e-mail services, LN supports collaborative work,
co-ordination of activities, and communication between different users, pursuing
common goals.
² LN, as an Intranet, is a central knowledge database with highly controlled security
access. It guarantees the security and con®dentiality of communication.
An application was developed within LN (for storage, enrichment and extraction) and
customised DE (puzzle creation and re®nement) in order to support the puzzle approach.
The resultingprototype supports the methodpresentedhere andhas the followingcapabilities.
1. PUZZLEallows a company to keep all employees informed of crucial events on a frequent
basis. It provides rapid storage and easy access to timely information, thereby reducing the
search time.
2. PUZZLEallows capture of users' knowledge fromevery branch of the organisation during
supply, enrichment of weak signs, or creation of puzzles. Users can access all information
and interpret it from different views. The prototype visualises information elements and
their enrichment in order to keep track of all persons who have enriched weak signs.
Therefore, it can be seen as a knowledge management system.
3. During the creation of puzzles, the prototype provides an environment to create multiple
arrangement of puzzles. Information mobility, link affectation, colours to differentiate
ideas, and a zoom function to focus on particular information elements are available.
4. During analysis of a puzzle, the prototype allows comments to be made and the author's
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 143
name and puzzle creation date to be stored in order to be able to compare different puzzles
created over a period of time.
5. Finally, PUZZLEcan be used as a working agenda for discussions when interpreting weak
signs. The output of such discussions is a puzzle map that is relevant, timely and contains
actionable intelligence that the decision makers may create.
When a manager from enterprise X (us) raises a question about a speci®c actor Y (i.e. a
competitor), information elements related to Y will be activated on the computer screen
according to criteria used to store data (see Section 4.2). The manager connects the informa-
tion elements by looking at their enrichment. In case some dif®culties occur, accessing the
help of the prototype provides de®nitions of links and examples. The manager then creates a
®rst puzzle about Y. If it stimulates his reasoning, it can be printed and individually used.
Otherwise, it can support collective discussions to derive other actions. Alternatively, the
puzzle can be cancelled and a newpuzzle generated using another idea. This procedure can be
iterated as many times as necessary. Fig. 3 illustrates an example of a puzzle concentrated on
Y and the theme the policy of Y toward R and D.
Link affectation is made by the authors using the prototype.
This puzzle could have been created by company X who puts Y under surveillance. The
elements of such a puzzle are connected by different links based on reasoning and
assessment of X. With the same information elements, it is possible to create several
puzzles until the most meaningful one is produced, or it suggests more interrogations
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 144
Fig. 3. Example of a Puzzle.
and hypotheses. This puzzle is dynamic since it can be modi®ed according to addition of
new weak signs. The analysis of that puzzle can lead to the following:
The expected event to be realised is Y may become a threat for X. Therefore, based upon
the sixth and seventh elements, the following hypothesis can be inferred: Y is becoming a
competitor of enterprise X. Taking this hypothesis into consideration, the following
actions should be undertaken by X:
² keep Y under surveillance and track it in order to discover why it is becoming a
competitor. What are its future strategies?
² de®ne the information sources for collecting more information about Y's actions and
assign persons to track the information,
² check the accuracy of: X is looking for an alliance with competitor Z, and evaluate the
impact of such an alliance on X,
² cross check the ®rst and second elements because they are in opposition.
In addition to its capabilities, the prototype suffers from some limitations. Its composi-
tion of two systems can be constraint. Weak signs are stored and enriched in LN before
they are exported to DE where puzzles are created and analysed. In addition, the prototype
does not automatically perform all these previous tasks. Information push for enrichment
and discussion, and puzzle creation through link affectation, are manual tasks. For instance
after a user accesses information, he can enrich it before sending it to other users. Once, a
user has completed the enrichment, the prototype integrates the views of all others users
which become visible to all. Actually PUZZLE allows collective enrichment, but it does
not support collective creation of puzzles. Finally, the prototype requires the assistance of
a facilitator.
6. Testing and results
This prototype is already operational and has been validated using several tests. Three
dif®culties have been highlighted (McCusker, 1992): disagreement over what to measure,
bene®t not easy to perceive, and bene®t not easy to quantify. Because of these issues, the
validation was based on two criteria:
1. usefulness of PUZZLE; this criteria refers to how PUZZLE (as a method and prototype)
enhances and supports the process of BI and the interpretation of weak signs.
2. ease of use of the prototype; this criteria refers to how practical and easy to comprehend
the concepts in PUZZLE are.
6.1. Subject and testing procedures
These tests were performed between 1997 and 2000 in France and the Netherlands,
using four methods.
First, in-depth interviews were held during several demonstrations of the prototype
(in France and the Netherlands). This validation has involved many participants from
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 145
companies who had been introduced to PUZZLE through the use of examples during a
three day exhibition. Interviews lasted between a half and one hour.
Second, the prototype was demonstrated during several initiation days, to eight
professors (three in the Netherlands and ®ve in France) with two consultants in France
and with six senior managers from French companies (four from an electronics company
and two from an insurance company). After being introduced to BI and PUZZLE, a
demonstration and interviews lasted 1/2 day.
Third, learning sessions were held with four senior managers from a Small and Medium
Sized Enterprise that develops electrical components. PUZZLE had been in use for two
months. With this third method, data collection followed a collective learning process
of four stages (Davis and Olson, 1985): basic understanding, individual learning,
recommendations, and tests.
Fourth, several tests were carried out with twenty MBA students from two Dutch
universities who were given a half a day introduction to PUZZLE. The tests were not
part of any course curriculum, nor did students receive academic credits for their
participation and no students had studied BI. After being introduced to PUZZLE, each
student was asked to create a puzzle (30 min) assisted by comments and advice from a
facilitator. Then, they were asked to create a puzzle collectively (30 min).
6.2. Data collection
To decrease researcher generated bias during the tests, two researchers were present,
one to operate PUZZLE and the other to collect data from direct observations. These
included the appropriateness of concepts (e.g. weak signs and links), assessment of
PUZZLE (identi®cation of strengths and weaknesses, adequacy of the conceptual
model compared with the empirical situation, perceived activity of puzzle creation).
In addition, participants who tested the prototype were surveyed in order to determine
their reactions.
The four tests demonstrated advantages and disadvantages of PUZZLE and opened new
perspectives.
6.3. Do users perceive PUZZLE as useful?
Among favourable reactions recorded were the following:
First, PUZZLE is seen as a structuring method that formalises heuristic reasoning,
clari®es BI and helps to perform it. All participants agreed that ªit is important for
managers to understand how to perform BI based upon weak signs which is completely
different from illegal, unethical economic espionageº. Also, all agreed that interpreting
weak signs in order to transform them into intelligence is both important and necessary
before becoming actionable information.
Second, PUZZLE allows better understanding of the concept `weak signs' and orients
the environmental scanning. It shows where BI bottlenecks are in order to propose
improvements ªcurrently, a number of departments across the company collect competi-
tive information on an ad-hoc basis, and very little of it is ever turned into useful intelli-
gence. Rather than anticipating what our competitors are going to do, we found ourselves
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 146
reading photocopied snippets of what they have already doneº, commented another
participant.
Third, PUZZLE is an organisational tool to support the co-ordination of scanning, helps
selecting weak signs and interpreting activities. As a consequence, the informal activities
of scanning currently operating within companies could be reduced. ªPUZZLE avoids
overloading managers and executives by selecting only crucial pieces of informationº,
commented one participant.
Fourth, the cognitive process of individuals is heuristic. Throughout the observations, it
seems that test participants of PUZZLE follow heuristic reasoning and not algorithmic: an
idea emerges, that launches others, then other information elements are searched, and the
process is iterated during creation of puzzles, hypotheses and commentaries. As indivi-
duals scan the external environment, the new information and interpretation they acquire
in¯uence their perception of environmental changes, which in turn affects what they will
perceive in their environment and how they will scan it. That ®nding suggests that BI is a
feedback process and that interpretation activities are close to that of the targeting and
selecting of weak signs.
Fifth, Puzzle stimulates interest and helps the interpretation of weak signs based solely
upon cognition ªYou give experts a tool to go further in reaction and in re¯ection. With
PUZZLE we are concentrated more on the cognition of experts. This is a very original
work which is different from most published work or existing tools like Datamining, for
exampleº commented a participant. This suggest that interpreting weak signs is much
concerned with human activity and as with technology.
Based on the positive reactions, we can assume that test participants perceive PUZZLE
as being useful for interpreting weak signs and for improving BI.
6.4. Do users perceive puzzle as easy to use?
In addition to the previous favourable reactions, participants also perceive PUZZLE as a
helpful tool that encourages communication and discussion. They also agreed that learning
BI skills could contribute to enhancing and improving career opportunities, students'
creativity, and could help students to learn real-world skills.
Despite the positive above reaction, PUZZLE shows some weaknesses. Even though
participants recognised the value of connecting weak signs to generate working hypoth-
eses, they perceived this activity to be dif®cult. They especially perceived this activity
more dif®cult when puzzles are created individually than collectively. According to other
test participants, puzzles are not suitable when many different information elements have
to be interpreted.
In addition, it was suggested that improving the collective method could give richer
weak signs, de®ne a typology of weak signs to facilitate the puzzles creation, and
develop a library of puzzles in order to facilitate their management. Finally, other test
participants thought that PUZZLE requires creative persons for interpreting weak
signs.
Based upon the above reactions, we can infer that PUZZLE is an easy to use educational
tool and the proposed concepts seemed to be understandable. However the proposed
concepts need further improvement and re®nements (Table 1).
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 147
7. Discussion and conclusions
This paper has presented a SBIS, based on exploratory research, to support managers in
practice to better handling weak signs and to ensure success of the business strategy. This
is an elusive issue related to knowledge management that has received little academic
attention.
This paper brings three main contributions: conceptual contribution in the form of a
creative method to support interpreting weak signs, design contribution in the form of
computer tool to support this method, and empirical contribution for understanding the
process of interpreting weak signs.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 148
Table 1
Summary of ®ndings
Criteria Expected
contribution
Findings
Perceived value of
PUZZLE to BI and
interpreting weak signs
BI process ± PUZZLE clari®es the content of BI
± PUZZLE fully supports the BI process (targeting, tracking, and
interpretation of weak signs)
± BI is a feedback process
± Puzzle is method to initiate performing BI activities
± PUZZLE helps selecting of weak signs
Interpreting
weak signs
± PUZZLE is seen as a structuring method that formalises heuristic
reasoning
± Transforming weak signs into intelligence is important and
necessary before becoming actionable information
± PUZZLE allows better understanding the concept of `weak
signs' and orients environmental scanning
± The cognitive process of individuals during weak sign
interpretation appears heuristic
± PUZZLE stimulates interest and helps the interpretation of weak
signs based solely upon cognition
± Interpretation activities are close to that of targeting and
selecting of weak signs
± PUZZLE is an organisational tool to support the co-ordination of
scanning and interpreting activities
Ease of use of PUZZLE Puzzle in
operation
± PUZZLE is a helpful tool for training and for quick learning of
BI and the interpretation of weak signs
± PUZZLE contributes to improving users' and students' BI skills,
their creativity, and their career opportunities
± PUZZLE helps students to learn real-world skills
± PUZZLE facilitates communication and discussion
Weakness
of PUZZLE
± Connecting weak signs is a dif®cult individual task but less so
when collectively undertaken
± Puzzles are not suitable when many information elements have
to be interpreted.
± PUZZLE requires creative persons for interpreting weak signs
These ®ndings have implications for both research and practice. From a practical
perspective, our ®ndings encourage the use of PUZZLE by managers to perform the
process of BI. The use of a tracking form, based on `actor/theme/weak sign/ personnel
enrichment/link, transformations', continuous discussion and creation of puzzles, are all
suitable for company's use. That form constitutes a shared language for understanding BI
and that will facilitate its use in companies. PUZZLE can also be a viable training method
for managers who want to learn BI process quickly and effectively.
From a research perspective this paper supports previous researches and adds new
®ndings. This paper suggests interpreting weak signs as both a process and a
product. The process requires the transformation of seemingly unrelated information
elements into useful puzzles (product) using links in order to produce hypotheses
about potential opportunities and threats. We found that collective interpretation
produces more value than when individuals work alone. Convergence of individual
interpretations following discussions seems to be fast and easy. These ®ndings
support the previous results of El Sawy and Pauchant (1988). Moreover, this
study suggests interpreting weak signs and BI as complex processes in which indi-
viduals follow a heuristic cognitive process. The discovery of managers' dif®culties
in connecting weak signs was totally unexpected because it had never been high-
lighted before and brings new ®ndings to strategic management literature. This
®nding also contradicts most statements of previous researches (Daft and Weick,
1984; Martinsons, 1994; Attaway, 1998; Subramanian and IsHak, 1998; Freeman,
1999).
In addition, this research encourages the use of the engineering research methodology
proposed by Channel et al. (1997). This could be very helpful for researchers who want to
understand the application of SIS in an area where little underlying theory exists, or for
ill-structured problems. This methodology is very useful because of its ability to stimulate
interest, create new information, open new research directions, and to propose re®nement
of concepts.
Results of this study should be examined in light of its limitations. First, we do not
claim that this research is other than exploratory since the sample size was small, which
limits the ability to generalise the result beyond this sample. Second, we obtained data
from a single informant and that may introduce respondent bias and limit the generalisi-
bility of the results. However, as companies compete in a turbulent environment, the
results and their implications are not industry-speci®c. Third, limits of this research should
also be examined with the reliability of its results. This is concerned with the extent to
which our results are reproducible and consistent with those produced by other
approaches. This is a dif®cult question to answer in any research study, particularly one
as exploratory as ours. Our intent was to explore whether the PUZZLE approach could
provide insights into the important research question posed above, and not to engage in a
rigorous comparison of alternative BI processes and software.
Future research can take several directions. First this research suggests to continue
testing PUZZLE, through replication under different conditions and in other organisations,
in order to learn more about weak signs interpretation including: what are the frequency of
using and updating puzzles? For which enterprises the PUZZLE approach is more suita-
ble? What are the characteristics of those enterprises (size, activities, etc.)? Does the use of
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 149
the prototype result in more creative ideas and better interpretation of weak signs? Also
there is a need to apply PUZZLE to a speci®c case study such as an emerging technology,
like the web telephone.
Second, ®ndings about the dif®culty to connect weak signs to each other suggests new
research directions toward understanding the adequacy of organisational memory models
with PUZZLE. One model that is well-known and respected in this ®eld is that of Horst
Rittel's Issue Based Information Systems (IBIS) (Rittel and Kunz, 1970). The IBIS model
focuses on the articulation of key issues in the design problem. Each issue can have many
positions. A position is a statement or assertion that resolves the issue. Each of an issue's
positions, in turn, may involve one or more arguments that either support that position or
object to it. Thus each separate issue is the root of a (possibly empty) tree, with the
children of the issues being positions, and the children of the positions being arguments.
In addition, new issues raised during discussion may be posted at any time and linked into
the most appropriate nodes. Those postings serve as an organisational memory that not
only captures the ®nal decisions when trees are `closed,' but also shows the history of the
alternatives explored (Conklin and Begeman, 1988). This paper encourages future
research to apply this model with puzzle by creating a graphical web based IBIS that
allows all members of a company to read and post an issue (weak sign), positions
(enrichments and questions), and arguments (links) at any time using their web
connections. Those suggestions will help to build a complete and coherent theory around
PUZZLE for BI.
Acknowledgements
Without the support of the Centre des Etudes et de Recherches AppliqueÂes aÁ la Gestion,
Ecole SupeÂrieure des Affaires, Universite Pierre MendeÁs (France) and the Information
and Technology department at the Eindhoven Faculteit Technology Management
(The Nertherlands), this study would not have been possible. The authors gratefully
acknowledge the contributions made by these organizations, professors Omar El Sawy
and Humbert Lesca, the Editor-in-chief and anonymous reviewers.
References
Aguilar, F., 1967. Scanning the business environment. Macmillan, New York.
Ansoff, I., McDonnell, E., 1990. Implanting strategic management. 2nd ed. Prentice-Hall, Englewood Cliffs, NI.
Ansoff, H.I., 1975. Managing strategic surprise by response to weak signals. California Management Review.
XXVIII (2), 21±33.
Attaway, M.C., 1998. A Review of issues related to gathering and assessing competitive intelligence. American
Business Review 16 (1), 125±135.
Beal, R.M., 2000. Competing effectively: Environmental scanning, competitive strategy, and organisational
performance in small manufacturing ®rms. Journal of Small Business Management 38, 27±47.
Bostrom, R.P., Nagasundaram, M., 1998. Research in Creativity and GSS. Proceedings of the Thirty-First Hawaii
International Conference on System Sciences, January 6±9. vol. 6. IEEE Computer Society, Piscataway, NJ
ISBN 0-(8186)-8248-5, Best paper in minitrack, pp. 391±505
Channel, V., Lesca, H., Martinet, A.C., 1997. Vers une ingeÂnierie de la recherche en science de gestion. Revue
FrancËaise de gestion, november±december, 41±51.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 150
Choe, J-M., Lee, Y-H., Park, K.-C., 1998. The relationship model between the in¯uence factors and the strategic
applications of information systems. European Journal of Information Systems 7, 137±149.
Clarke, R., 1994. The path of development of Strategic Information Systems Theory. Available from http://
www.anu.edu.au/people/Roger.Clarke/SOS/StratISTh.html.
Conklin, J., Begeman, M., 1988. gIBIS: A Hypertext tool for Exploratory Policy Discussion. ACM Transactions
on Of®ce Information Systems 6, 4 October.
Couger, J.D., Higgins, L.F., McIntyre, S.C., 1993. (Un)structured creativity in information systems organizations.
MIS Quarterly December, 375±397.
Courtial, J.P., 1994. A coword analysis of scientometrics. Scientometrics 31, 251±260.
Cyert, R., March, J., 1963. A behavioral theory of the ®rm. Prentice-Hall, Englewood Cliffs, NJ.
Daft, R., Weick, K.E., 1984. Toward a model of organisation as interpretation systems. Academy of Management
Review 9, 284±296.
Davis, G.B., Olson, M., 1985. Management information systems: foundations, structure and development.
McGraw-Hill, New York.
Drucker, P.F., 1993. Managing in Turbulent Times. Reprint edition. Harper Collins Publishers, ISBN:
0887306160.
El Sawy, O., 1985. Personal information systems for strategic scanning in turbulent environments: can the CEO
go on-line? MIS Quarterly March, 53±60.
El Sawy, O.A., Pauchant, T., 1988. Triggers, templates and twitches in the tracking of emerging strategic issues.
Strategic Management Journal 9, 455±473.
Freeman, O., 1999. Competitor intelligence: information or intelligence. Business Information Review 16 (2),
71±77.
Futures Group, 1997. Ostriches and eagles 1997, Futures Group Inc. (August) available from http://www.tfg.com/
pubs/docs/0_Elll-97.html.
Galliers, R.D., 1991. Strategic Information Systems planning: myths reality and guidelines for successful
implementation. European Journal of Information Systems 1 (1), 55±64.
Gilad, T., Gilad, B., 1986. Business intelligence. The quiet revolution. Sloan Management Review 27, 53±62.
Groom, J.R., David, F.R., 2001. Competitive intelligence activity among small ®rms. SAM Advanced Manage-
ment Journal winter, 12±20.
Grove, A., 1999. Only the paranoid survive: How to Exploit the Crisis Points that Challenge Every Company. 1st
current edition, Bantam Books.
Hall, C., 2001. The intelligent puzzle. Competitive Intelligence Review 12 (4), 3±14.
Herring, J.P., 1998. What is intelligence analyst? Competitive intelligence Magazine 1 (2), 13±16.
Kanter, R.M., 1988. Change-Master Skills: What it takes to be creative. In: Kuhn, R.L. (Ed.). Handbook for
Creative and Innovative General Managers. McGraw-Hill, New York, NY, pp. 91±99.
Kelly, K., 1998. New rules for the new economy: ten ways the network economy is changing everything. Fourth
Estate, London.
Koneig, G., 1996. Management strategique, paradoxes, interactions et apprentissages. Nathan, Paris.
Lesca, H., Caron, M.L., 1995. Veille strateÂgique: creÂer une intelligence collective au sein de l'entrepise. Revue
FrancËaise de Gestion septembre±octobre, 58±68.
Lesca, H., 1994. Veille strategique pour le management strategique eÂtat de la question et axes de recherche.
Economies et societeÂs, seÂrie Science de Gestion 5 (20), 31±50.
Maccrimmon, K.R., Wagner, C., 1994. Stimulating ideas through creativity software. Management Science 40
(11), 1514±1532.
Martinsons, M.G., 1994. A strategic vision for managing business intelligence. Information Strategy Spring, 17±
33.
McCann, J., Gomez-Meija, L., 1992. Going on-line in the environmental scanning process. IEEE Transaction on
Engineering Management 39 (4), 394±399.
McCusker, I.C., 1992. IT effectivenessÐWhat does management need to know? The EDP Auditor Journal 3, 25±
29.
Meyer, A.D., 1991. Visual data in organizational research. Organisation Science 2 (2), 218±236.
Miller, G.A., 1956. The magical number seven, plus or minus two: some limits on our capacity for processing
information. Psychological Review 63 (2), 81±97.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 151
Moore, G., McKenna, R., 1999. Crossing the chasm: marketing and selling high-tech products to mainstream
customers. Rev ed. Perennial (Harper Collins) publisher, ISBN: 0066620023.
Nolan, J.A., 1999. It is the third millennium: do you know where your competitor is? Journal of Business Strategy
November/December, 11±15.
Norman, 1982. Learning and Memory. W.H. Freeman, San Francisco, CA.
Piaget, J., 1996. EpisteÂmologie geÂneÂtique. Press Universitaire de France.
Plsek, P.E., 1996. Working paper: models for the creative process. available from http://www.directedCreativity.
com/pages/WPModels.html.
Rittel, H., Kunz, W., 1970. Working paper. Issues as Elements of Information Systems. 131. Institute of Urban
and Regional Development, University of California, Berkeley, CA.
Rouibah, K., 1998. Puzzle: a business intelligence tool for the interpretation of anticipatory and fragmentary weak
signs. PhD Dissertation, Ecole SupeÂrieure des Affaires, Universite Pierre MendeÁs France, Grenoble, France.
Rouibah, K., Lesca, H., Malek, S., 1997. Business intelligence. A processing heuristic for weak signal. 15th
Annual International Conference of AoM/IAoM, Montreal, QueÂbec, Canada, 6±9 aouÃt 15 (1), 97±106.
Sabherwal, R., King, W., 1995. An emprical taxonomy of the decision-making process concerning strategic
applications of information systems. Journal of Management Information Systems 11 (4), 177±214.
Schneider, S.C., Meyer, A.D., 1991. Interpreting and responding to strategic issues: the impact of national culture.
Strategic Management Journal (12), 307±320.
Simon, H., 1960. The new science of management decision. Prentice-Hall, Englewood Cliffs.
Stenberg, R.J., 1988. The nature of creativity. Cambridge University Press, New York.
Subramanian, R., IsHak, S., 1998. Competititor analysis practices of US companies: an empirical investigation.
MIR Management International Review 3, 7±23.
Thomas, J.B., McDaniel, B.R., 1990. Interpreting strategic issues: effects of strategy and the information-proces-
sing structure of top management teams. Academy of Management Journal 33 (2), 286±306.
Valette, F., 1993. Le concept de Puzzle: coeur du processus d'eÂcoute prospective de l'environnement de l'en-
treprise. ThDoc ESA, France.
Wagner, J.A., Gooding, R.Z., 1997. Equivocal information and attribution: an investigation of patterns of
managerial sense making. Strategic Management Journal 18 (4), 275±286.
Walls, J.G., Widmeyer, G.R., El Sawy, O.A., 1992. Building an information system design theory for vigilant
EIS. Information System Research 3 (1), 36±59.
Wiseman, C., 1988. Strategic Information Systems. Irwin, Home-Wood, IL.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 152
doc_367857659.pdf
A Concept and Prototype for Linking Business Intelligence To Business Strategy
PUZZLE: a concept and prototype for linking
business intelligence to business strategy
Kamel Rouibah
*
, Samia Ould-ali
Department of Quantitative Methods and Information Systems, College of Business Administration,
P.O. Box 5486 Safat 13055, Kuwait
Received 29 November 1999; revised paper accepted 18 March 2002
Abstract
Business intelligence (BI) is a strategic approach for systematically targeting, tracking,
communicating and transforming relevant weak signs
1
into actionable information on which stra-
tegic decision-making is based. Despite the increasing importance of BI, there is little underlying
theoretical work, which directly can guide the interpretation of ambiguous weak signs. This paper
gives an insight into the issue through a new strategic business intelligence system called PUZZLE.
We describe this system and validate it by designing a prototype, test the system using in-depth
interviews, and hold learning sessions in order to further knowledge about BI. The main results from
tests show that: interpreting weak signs is potentially important for senior managers, consultants, and
researchers; interpretation can be achieved gradually by bringing the weak signs together using a
tracking form based upon the concept of actor/theme/weak signs/enrichment /links; interpreting
weak signs is a complex process of establishing links between the weak signs. Final results show
that the individual cognitive process appears heuristic when interpreting weak signs. Implications for
strategic management practice and research are addressed. q 2002 Elsevier Science B.V. All rights
reserved.
Keywords: Business intelligence; Weak signs; Interpreting weak signs; Fast response management; Strategic
information system; Strategic business intelligence system; Creativity; Ill-structured problem; Research engi-
neering; Exploratory research; Prototyping
Journal of Strategic Information Systems 11 (2002) 133±152
0963-8687/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved.
PII: S0963-8687(02)00005-7
www.elsevier.com/locate/jsis
* Corresponding author. Tel.: 1965-484-2671x8445; fax: 1965-254-9408.
E-mail address: [email protected] (K. Rouibah).
1
The authors use the concept of weak sign instead of `weak signal' as proposed by Ansoff, 1975 because they
feel the word `signal' implies greater quantitative measurability. However, authors continue to use it for the same
purpose.
1. From business intelligence to interpreting weak signs
Companies are evolving in turbulent and equivocal environments (Drucker, 1993;
Kelly, 1998; Grove, 1999). This requires companies to be alert and watchful for the
detection of weak signals (Ansoff, 1975) and discontinuities about emerging threats and
opportunities and to initiate further probing based on such detection (Walls and
Widmeyer, 1992. In such environments, business intelligence (BI) is surfacing to deal
with the large volume of information available but which are often misleading, inaccurate
and untimely (Martinsons, 1994; Futures Group, 1997; Attaway, 1998; Herring, 1998;
Freeman, 1999; Groom and David, 2001). The main crucial question raised by companies
in such environments is how to exploit these information elements to grasp opportunities,
and avoid surprises when discontinuities occur (Grove, 1999; Moore and McKenna,
1999). This is the reason that companies need to have a well analysed, designed, and
developed strategic business intelligence system (SBIS) (Martinsons, 1994). The emphasis
here is on information systems that enhance strategic decision-making and that support the
competitive strategy of an organisation (Wiseman, 1988). Much has been written on
environmental scanning systems since Aguilar, Ansoff and Porter (e.g. Beal, 2000).
However, it seems that the growing uncertainty of business environments still raises the
need for an ef®cient SBIS to support scanning and interpreting information so that
valuable intelligence may be delivered to senior managers. This paper develops a new
SBIS and its implementation as a computer system. It is oriented to improve the BI and to
ensure success of the business strategy. The meaning of BI emphasised in this paper,
termed in French as veille strateÂgique, is considered as a systematic approach by which
a company keeps itself vigilant and aware of developments and early warning signs in its
external environment in order to anticipate business opportunities or threats. The external
environment includes all factors and events outside the company that can affect its
performance. Designing a successful SBIS requires an understanding of the relationship
between the BI process and weak signs. There are several similar variants of BI processes
(Martinsons, 1994; Attaway, 1998; Nolan, 1999). Our intent is not to describe in detail
those approaches but only to show the link to weak signs. The description of these similar
processes is beyond the scope of this paper. Among these, one is certi®ed ISO 9001 and is
mainly oriented toward weak signs management. This process is explained below.
1.1. Focus of this research
We consider the BI process as cyclical, involving ®ve phases (Fig. 1, adapted from
Lesca, 1994).
The ®rst phase `targeting' consists of bounding the surveillance of the company's
environment to set tracking priorities. The second phase consists of organising tracking
and selecting the crucial weak signs. The third phase consists of routing the weak signs
collected from outside to inside the organisation. The fourth phase `interpreting' consists
of transforming the collected information into actionable intelligence. If interpretation is
signi®cant, actions can be taken in phase 5. Otherwise, information search has to be re®ned
in a more speci®c way (return to phase 2) if information is imprecise or; (2) the boundary
(target) has to be rede®ned (return to phase 1) if it is too large. Of these ®ve phases, the
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 134
fourthÐinterpretingÐis the most important and most dif®cult (Rouibah, 1998; Attaway,
1998; Subramanian and IsHak, 1998). Inadequate transformation leads to failures in using
BI (Martinsons, 1994; Lesca and Caron, 1995).
Much has been written about the characteristics associated with events as threats or
opportunities, their consequences and people's reactions (Thomas and McDaniel, 1990;
Schneider and Meyer, 1991; Wagner and Gooding, 1997). However, researchers have not
directly examined how people interpret weak signs in order to identify whether they turn
into opportunities or threats even though interpreting weak signs is recognised as an
important research topic for strategic decision-making (El Sawy and Pauchant, 1988;
Thomas and McDaniel, 1990; Ansoff and McDonnell, 1990; Martinsons, 1994; Freeman,
1999). In addition, despite the need for strategic intelligence delivery being widely
recognised (Attaway, 1998; Herring, 1998; Nolan, 1999; Freeman, 1999; Groom and
David, 2001), the problem of interpreting weak signs has not received much attention
in the academic world.
While providing insight into the problem was an operational objective of this study,
there were also important theoretical and practical questions that had to be asked:
1. How to interpret weak signs in order to create meaningful maps of environmental
changes useful to managers?
2. In what format should weak signs be presented?
3. How can such maps be created?
From a practical perspective, answers were sought to discover what dif®culties and
operational questions are associated with the interpretation of weak signs.
Hopefully, this paper provides some insight into those issues. The remaining segments
of the paper are organised as follows. In the next segment dif®culties of interpreting weak
signs and research methodology are discussed. In the third segment, relevant and current
literature on the subject of this paper is reviewed. In the fourth segment of the paper, a new
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 135
Fig. 1. Phases of business intelligence.
prescriptive approach to deal with weak sign and the speci®cation of the computer tool's
functionality will be introduced. The ®fth segment presents prototype capabilities and
limitation. The sixth segment of the paper deals with lessons learned from validation of
the approach. The concluding segment of the paper offers the implications of our ®ndings
for both academics and practitioners.
2. Work context and research methodology
Before describing the research methodology used, Sections 2.1±2.3 discuss the context
of weak signs and the concept of strategic information systems (SIS).
Two reasons explain the dif®culty in interpreting weak signs: the position of such
information in strategic decision-making, and the nature of the information involved.
2.1. The position of weak signs in decision making
Two generic modes of scanning can be acknowledged: reactive and proactive. In the
reactive mode of scanning (Cyert and March, 1963) search is stimulated by a problem and
directed towards supporting strategic decisions. Its bene®ts are readily apparent to
managers because the information collected could be directly integrated into strategic
decision-making. However, the proactive mode of scanning or surveillance is exploratory
and not directed to any prede®ned problem (Aguilar, 1967). In the latter case, the required
information cannot be de®ned in advance. Its bene®ts are less easily measured. Therefore,
monitoring activities are not fully integrated into the decision-making process (Lesca and
Caron, 1995), senior managers continue to make poor decisions (Martinsons, 1994), and
weak signs are ignored (Rouibah, 1998). Weak signs associated with the proactive mode
may lead to a change in business strategy and therefore will require transformation into
actionable intelligence before the information becomes usable. The proactive mode is the
main concern of this paper.
2.2. The nature of information involved
The particular characteristics of weak signs make them very dif®cult to manage.
According to El Sawy (1985) and Rouibah et al. (1997), a weak sign is:
² anticipatory: it informs about changes when they begin to occur,
² uncertain: it concerns events that have not yet been realised,
² ambiguous: its content is usually uncertain or could be deliberately contaminated or
distorted (for example by a competitor),
² fragmentary: each information element taken alone is insigni®cant; however,
signi®cance increases gradually when combined with other weak signs,
² dynamic: it evolves over time,
² cyclical: it has a complex life cycle from growing to declining that varies in its duration
and signi®cance (Ansoff, 1975; McCann and Gomez-Meija, 1992),
² qualitative: in most cases, numbers are not involved and information may be available
in other forms such as written, verbal, or visual images.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 136
Weak signs are thus subject to perception and interpretation, and multiple meanings are
possible. In order to avoid any confusion in the use of the word signal, we prefer to call
such information a weak sign. Here, we de®ne weak signs as uncertain and fragmented
information about developments and trends. These have not been completely realised, or
they have potential consequences, or are perceived to have a signi®cant impact on
organisational performance, either as threats or opportunities.
Taking into account the previous explanation, it is clear that quantitative and archival
information available in on-line computerised databases (OCDBs) (McCann and
Gomez-Meija, 1992) are out of the scope of this paper. They can help to track changes
in world markets and to decrease the time needed to discover trends. OCDBs rely on
information that can be codi®ed and entered textually or numerically. They are used in
combination with well-known scientometrics and bibliometric techniques (e.g. see
Courtial, 1994). These are more concerned with how often archived information is
consulted, and are very often used to perform trend analysis in research (e.g. assessment
of laboratory results, comparison between laboratories). Nevertheless, these tools are
inappropriate to assess unwritten information, which come from personal sources as
some weak signs may.
Based on the previous dif®culties, it can be assumed that a manager's attitude towards
BI would be more effective if a SBIS were developed to assist with a better handling of
weak signs. Section 2.3 explains what we mean by a SBIS.
2.3. Strategic information systems for business intelligence (SBIS)
Strategic information systems (SIS) are systems which contribute signi®cantly to an
organisation's performance and, consequently, to play a major role in its competitive
strategy. A SIS is de®ned as a system having a profound effect on business success by
in¯uencing or shaping a company's strategy or by playing a direct role in the implementa-
tion of that strategy (Sabherwal and King, 1995; Galliers, 1991). An alternative interpreta-
tion of SIS suggests that it is not necessarily a particular information system, but rather a
combination of those parts of an organisation's cluster of information systems, which feed
its strategy planning processes (Clarke, 1994). Among factors that in¯uence the SIS
application, external factors (such as environmental competition, uncertainty and external
needs) are driving forces for any SIS application (Choe et al., 1998). A SIS-oriented
toward external changes helps an organisation to remain competitive and proactive.
As a result, all components of information systems, which support all or part of BI
phases (Fig. 1) are considered to be SIS. If the above is agreed, then we consider a SBIS as
SIS oriented to support company's BI activities.
2.4. Research methodology
Interpreting weak signs is a real problem encountered by many managers in different
organisations that is closely related to the strategic decision-making process. This process
is recognised as ill-structured and not adequately understood (Walls and Widmeyer, 1992;
Schneider and Meyer, 1991; Channel et al., 1997). While the current situation is known
(dif®culty of the interpreting weak signs), the target or ideal situation that informs
managers about potential threats or opportunities is unknown. Surveys (collecting large
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 137
amounts of data followed by testing hypotheses) are not appropriate for helping managers
in this situation. Therefore, the chosen methodology used to perform this research has been
called `engineering research' (Channel et al., 1997). It is an on-site methodology useful
for expanding knowledge about complex and ill-structured processes, such as interpreting
weak signs, and helping managers to progress.
Characteristics of this research methodology are as follows:
1. Engineering activities: the researcher is not an observer, but is a part of the problem
solving process. He acts as an engineer who develops a conceptual model, builds a
prototype, acts as a facilitator and evaluates his construction on site in the organisation.
2. Exploratory research: due to the lack of structure underlying the problem addressed, it
can only be approached by exploratory research based on trial and error, in modelling,
building, testing and evaluating.
3. Creation of procedural knowledge: it refers to the knowledge that helps to act and to
reason (the knowledge of what to do next). This takes the form of a method, a guideline
or a prototype.
The main results expected from this methodology fall into the following:
1. Ability to stimulate interest. Most senior managers or executives needing to be
interviewed are not always able to express clearly either their dif®culties or their
requirements. Therefore, it is useful to support them with `something' in the form of
a prototype in order to attract them and motivate their meeting with researchers.
2. Prototype. This can be used as a support for training and learning of users who will
learn faster by doing rather than by talking. We improve the prototype using users'
suggestions.
3. Creation of new information. Testing the prototype enables the collection of users'
reactions and relevant observations, which would be out of reach without the prototype
or when traditional methods such as surveys are used.
4. Open new research directions. Negative users' reactions during tests become new
research directions and new concepts to develop. The feedback (data collected during
tests with the end users) can be re-used to improve the interpreting process in order to
support managers' needs and contribute to more theoretical knowledge in the ®eld of
BI.
5. Re®nement of concepts. Based upon users' reactions and suggestions, we go back to
the concepts (artefact) we developed in order to further knowledge and propose
enhancement and re®nement.
Section 3 speci®es the appropriate literature with regard to advantages and limitations to
the model developed.
3. Literature review
Several researchers admit that the information collected must be transformed into
actionable intelligence before it becomes usable (Martinsons, 1994; Lesca and Caron,
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 138
1995; Herring, 1998; Koneig, 1996; Attaway, 1998). Moreover, to achieve such an
objective, authors propose to integrate unrelated pieces of information in order to create
a holistic picture (Valette, 1993; Martinsons, 1994; Lesca and Caron, 1995; Koneig, 1996;
Attaway, 1998; Subramanian and IsHak, 1998; Freeman, 1999).
However this transformation has received little attention. Only two methods 3T (El
Sawy and Pauchant, 1988) and PUZZLE (Valette, 1993), to our knowledge, have
addressed the problem. According to those authors, interpreting weak signs can be done
based on a tracking form composed of `actor/theme
2
/information'. Pieces of information,
related to a theme and a speci®c actor (e.g. a competitor), can be pieced together to get a
mental map `puzzle'. The term `puzzle' has been used by several researchers to create an
analogy between interpreting weak signs and reconstruction of a puzzle (see Gilad and
Gilad, 1986; Valette, 1993; Lesca and Caron, 1995; Attaway, 1998; Hall, 2001). Valette
suggests using PUZZLE to detect signi®cant environmental changes and to orient the
scanning. El Sawy and Pauchant found collective interpretation produces more value
than when individuals work alone. They also found that a shift in the interpretation is
generated through the perception of new weak signs or the occurrence of new learning and
ideas. Moreover, a change in a cognitive map takes place through cognitive functions,
which are invariant (content independent), and thus, can be applied to a variety of tasks
and situations. However, these cognitive functions have not been studied and implemented
as a guideline. We propose to reuse the idea of integrating bits of information together in
the form of puzzles and to build a new methodÐaccompanied by a prototypeÐin line
with the two previous ones, which we continue to refer to as PUZZLE. In order to
complete the understanding of weak signs and the cognitive functions that occur during
the interpretation we turn to creativity and learning.
According to several authors output and process are key concepts in creativity (e.g.
Couger et al., 1993; MacCrimmon and Wagner, 1994; Bostrom and Nagasundaram, 1998).
The output is built up gradually by successive alteration (Plsek, 1996). One process that is
common to many theoretical contributions is making connections or associations (see
Stenberg, 1988; Kanter, 1988; MacCrimmon and Wagner, 1994). This process is one of
the main focus of this paper; it refers to the creation of new ideas through relationships
between existing ideas. In our model, such connections can be triggered by linking pieces
of information together in order to infer new information from adjacent ones; and to
generate hypotheses that must be validated quickly. This idea is derived from the
kaleidoscopic thinking of Kanter (1988). It allows the creation of arrangements by
connecting fragmentary information elements until new patterns and actions can be
generated. Such idea is also in line with Herring's view of intelligence as a process in
which information is subject to systematic examination and determination of signi®cant
relationships (Herring, 1998). Weak signs are not all alike since they probably come from
different sources. Thus, they can support or contradict each other, or one weak sign may be
the cause, the consequence, or the result of another weak sign. The process of making
connections was implemented by connecting weak signs through a typology of links:
causality, con®rmation, and contradiction.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 139
2
The term `theme' is used in cognitive psychology and stresses a dynamic description rather than a static point
of view (El Sawy and Pauchant, 1988).
The limited rationality theory (Simon, 1960) gives insight on the human interpretation
of issues. According to this theory, the human brain has limited capacities for interpreting
information and it is in¯uenced by the nature of the information. Miller (1956) showed that
the maximum number of chunks (weak signs in our model) that could be remembered and
dealt with at one time in short-term memory, is 7 ^2. Based on Miller's ®ndings, Simon
proposed a reasoning mode that is heuristic. This requires a manager to build partial
environmental maps. These help him to act even though they are incomplete. In addition,
according to research in cognitive theory, individuals treat information better and faster
when it is presented visually (Meyer, 1991). Therefore, we have chosen to represent the
interpreted information as a graphical map, that we continue to refer to as puzzle. Puzzles
are composed of weak signs and their links. During the creation of a cognitive puzzle,
different modi®cations can re¯ect what has been learned. Piaget (1996) and Norman
(1982) identi®ed the following modes of learning: assimilation where new information
is assimilated into an old cognitive map; accommodation where old cognitive maps are
modi®ed when new information does not ®t the old cognitive map; and structuring when
new cognitive maps are formed. Then, it can be inferred that elements of a puzzle will be
subject to addition, deletion, and mergers.
Based on that previous observations it should now be clear that there is a very intimate
and dynamic relationship between interpretation of weak signs and creativity.
The following method was developed from the previous observations.
4. Proposed method
We de®ne the interpretation of weak signs as both a process and product:
² a process by which an individual or group of individuals transform weak signs into
meaningful maps, even partial, about how a company's present and future business
environment may be impacted,
² an end product, that is the output of that process.
Fig. 2 shows the conceptual model for the interpretation.
4.1. Phases of the method
We propose a method based upon a creative process of seven steps.
4.1.1. Step 1Ðinput to the method
One way to establish boundaries of the environment is to limit the surveillance effort to
(a) the speci®c targets (actors
3
/theme) and; (b) to key strategic information sources. Weak
signs collected according to targets are input to the method. These information elements
are in their early stage and have been collected according to surveillance mode.
Interpreters are non-experts.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 140
3
Actors are all factors and events outside the company that affect its performance, e.g. suppliers, competitors.
4.1.2. Step 2Ðenrichment of the weak signs
All employees of a company could be a potential source of valuable intelligence. A
single information element may lead to signi®cantly different interpretations based on its
context. One might enrich an element in different ways: assess its content using certain
criteria (surprise, validity, relevance, timeliness, completeness, accuracy), provide
continuations or contradictions to the existing information, and to complete missing
information. The value of a weak sign is enhanced when all the views are integrated.
4.1.3. Step 3Ðcategorisation of weak signs
Once the content of a weak sign has been evaluated, it can be categorised and classi®ed
according to themes. These can be identi®ed during a learning session and bounded by an
initial de®nition. A theme is a subject of interest that informs about an external actor for
example the R&D policy of a competitor.
4.1.4. Step 4Ðcreation of puzzles
This step consists of selecting certain information elements from the themes in order to
elaborate puzzles. The construction of a puzzle is a real act of creativity because it consists
of linking weak signs together in such a manner that fragmentary elements become an
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 141
Fig. 2. Framework of weak signs interpretation.
intelligible map, even if incomplete. A puzzle is a visual map, focused on a speci®c actor,
where its nodes are small sentences (phrases) corresponding to weak signs; and edges are
reasoning links (con®rmation, contradiction, causality) which connect different nodes.
Puzzles are useful to perceive a change when they are compared at different times.
4.1.5. Step 5Ðperception of modi®cation
Different transformations may occur when examining a initial puzzle called G
0
:
1. delete information in G
0
, if it is obsolete;
2. subdivide rich information of G
0
;
3. merge two information elements of G
0
into one element when they are identical or one
of them is redundant;
4. add a link between two information elements of G
0
i.e. one has been identi®ed as a new
link such as a continuation;
5. delete a link between two information elements of G
0
4.1.6. Step 6Ðre®nement of puzzles
Each puzzle must be ¯exible, and continually restructured as new weak signs are
collected. Examination of a puzzle based upon new information could challenge the
existing information in contradiction, con®rmation, or continuation. Thus, other
supplementary transformations can take place in the puzzle G
0
, other than the ®ve
above: modi®cation of an information content; substitution of information (new
information replaces existing).
4.1.7. Step 7Ðproposal of actions
Going from atomised pieces of information to a re®ned puzzle helps users to reason.
Arguments and questions are required to conduct such reasoning process:
² Infer new information from the related and adjacent elements that needs immediate
validation.
² Provide progressive veri®cation of the coherence between information elements.
² Orient the scanning and tracking of new and missing information.
4.2. Functions needed for implementation
According to our model, creation of puzzles requires an appropriate computer tool that
is composed of a database and an ideas mapping tool to support cognitive mapping. While
the database will store, and retrieve information, and capture users' knowledge during
enrichment activities (pull for enrichment and push for discussion); the drawing graph will
support the creation and analysis of puzzles. Co-ordination and communication are also
necessary functions for collective interpretation.
A form will record the following for each information element:
² a title, this is a short anticipatory information about actions of an actor,
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 142
² a tracker that collects the information,
² a theme, to which information elements are related,
² an information source that generates information,
² each information element is enriched,
² each information element is evaluated in terms of importance/reliability so that differ-
ent representations can be generated according to their importance and reliability.
² each information has an editing date,an expired date and a date for the expected event,
² an end user in the organisation who needs the information.
To ensure the security and con®dentiality of the system, each user should have
identi®cation and privileges in storage, enrichment, extraction of information, and access
to a puzzle summary.
5. Prototype capabilities and limitations
The prototype developed, called PUZZLE, comprises Lotus Notes (LN) and Decision
Explorer (DE) (from Banxia). DE was chosen because it can be easily customised to
generate puzzle maps. On the other hand, LN was chosen for three main reasons.
² LN, as a technology platform, is able to create many applications. It contains various
sophisticated tools and an integrated macro language, which allows further
development and customisation of the applications.
² LN, as groupware, allows several users to work simultaneously on the same application.
Through its shared databases and e-mail services, LN supports collaborative work,
co-ordination of activities, and communication between different users, pursuing
common goals.
² LN, as an Intranet, is a central knowledge database with highly controlled security
access. It guarantees the security and con®dentiality of communication.
An application was developed within LN (for storage, enrichment and extraction) and
customised DE (puzzle creation and re®nement) in order to support the puzzle approach.
The resultingprototype supports the methodpresentedhere andhas the followingcapabilities.
1. PUZZLEallows a company to keep all employees informed of crucial events on a frequent
basis. It provides rapid storage and easy access to timely information, thereby reducing the
search time.
2. PUZZLEallows capture of users' knowledge fromevery branch of the organisation during
supply, enrichment of weak signs, or creation of puzzles. Users can access all information
and interpret it from different views. The prototype visualises information elements and
their enrichment in order to keep track of all persons who have enriched weak signs.
Therefore, it can be seen as a knowledge management system.
3. During the creation of puzzles, the prototype provides an environment to create multiple
arrangement of puzzles. Information mobility, link affectation, colours to differentiate
ideas, and a zoom function to focus on particular information elements are available.
4. During analysis of a puzzle, the prototype allows comments to be made and the author's
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 143
name and puzzle creation date to be stored in order to be able to compare different puzzles
created over a period of time.
5. Finally, PUZZLEcan be used as a working agenda for discussions when interpreting weak
signs. The output of such discussions is a puzzle map that is relevant, timely and contains
actionable intelligence that the decision makers may create.
When a manager from enterprise X (us) raises a question about a speci®c actor Y (i.e. a
competitor), information elements related to Y will be activated on the computer screen
according to criteria used to store data (see Section 4.2). The manager connects the informa-
tion elements by looking at their enrichment. In case some dif®culties occur, accessing the
help of the prototype provides de®nitions of links and examples. The manager then creates a
®rst puzzle about Y. If it stimulates his reasoning, it can be printed and individually used.
Otherwise, it can support collective discussions to derive other actions. Alternatively, the
puzzle can be cancelled and a newpuzzle generated using another idea. This procedure can be
iterated as many times as necessary. Fig. 3 illustrates an example of a puzzle concentrated on
Y and the theme the policy of Y toward R and D.
Link affectation is made by the authors using the prototype.
This puzzle could have been created by company X who puts Y under surveillance. The
elements of such a puzzle are connected by different links based on reasoning and
assessment of X. With the same information elements, it is possible to create several
puzzles until the most meaningful one is produced, or it suggests more interrogations
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 144
Fig. 3. Example of a Puzzle.
and hypotheses. This puzzle is dynamic since it can be modi®ed according to addition of
new weak signs. The analysis of that puzzle can lead to the following:
The expected event to be realised is Y may become a threat for X. Therefore, based upon
the sixth and seventh elements, the following hypothesis can be inferred: Y is becoming a
competitor of enterprise X. Taking this hypothesis into consideration, the following
actions should be undertaken by X:
² keep Y under surveillance and track it in order to discover why it is becoming a
competitor. What are its future strategies?
² de®ne the information sources for collecting more information about Y's actions and
assign persons to track the information,
² check the accuracy of: X is looking for an alliance with competitor Z, and evaluate the
impact of such an alliance on X,
² cross check the ®rst and second elements because they are in opposition.
In addition to its capabilities, the prototype suffers from some limitations. Its composi-
tion of two systems can be constraint. Weak signs are stored and enriched in LN before
they are exported to DE where puzzles are created and analysed. In addition, the prototype
does not automatically perform all these previous tasks. Information push for enrichment
and discussion, and puzzle creation through link affectation, are manual tasks. For instance
after a user accesses information, he can enrich it before sending it to other users. Once, a
user has completed the enrichment, the prototype integrates the views of all others users
which become visible to all. Actually PUZZLE allows collective enrichment, but it does
not support collective creation of puzzles. Finally, the prototype requires the assistance of
a facilitator.
6. Testing and results
This prototype is already operational and has been validated using several tests. Three
dif®culties have been highlighted (McCusker, 1992): disagreement over what to measure,
bene®t not easy to perceive, and bene®t not easy to quantify. Because of these issues, the
validation was based on two criteria:
1. usefulness of PUZZLE; this criteria refers to how PUZZLE (as a method and prototype)
enhances and supports the process of BI and the interpretation of weak signs.
2. ease of use of the prototype; this criteria refers to how practical and easy to comprehend
the concepts in PUZZLE are.
6.1. Subject and testing procedures
These tests were performed between 1997 and 2000 in France and the Netherlands,
using four methods.
First, in-depth interviews were held during several demonstrations of the prototype
(in France and the Netherlands). This validation has involved many participants from
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 145
companies who had been introduced to PUZZLE through the use of examples during a
three day exhibition. Interviews lasted between a half and one hour.
Second, the prototype was demonstrated during several initiation days, to eight
professors (three in the Netherlands and ®ve in France) with two consultants in France
and with six senior managers from French companies (four from an electronics company
and two from an insurance company). After being introduced to BI and PUZZLE, a
demonstration and interviews lasted 1/2 day.
Third, learning sessions were held with four senior managers from a Small and Medium
Sized Enterprise that develops electrical components. PUZZLE had been in use for two
months. With this third method, data collection followed a collective learning process
of four stages (Davis and Olson, 1985): basic understanding, individual learning,
recommendations, and tests.
Fourth, several tests were carried out with twenty MBA students from two Dutch
universities who were given a half a day introduction to PUZZLE. The tests were not
part of any course curriculum, nor did students receive academic credits for their
participation and no students had studied BI. After being introduced to PUZZLE, each
student was asked to create a puzzle (30 min) assisted by comments and advice from a
facilitator. Then, they were asked to create a puzzle collectively (30 min).
6.2. Data collection
To decrease researcher generated bias during the tests, two researchers were present,
one to operate PUZZLE and the other to collect data from direct observations. These
included the appropriateness of concepts (e.g. weak signs and links), assessment of
PUZZLE (identi®cation of strengths and weaknesses, adequacy of the conceptual
model compared with the empirical situation, perceived activity of puzzle creation).
In addition, participants who tested the prototype were surveyed in order to determine
their reactions.
The four tests demonstrated advantages and disadvantages of PUZZLE and opened new
perspectives.
6.3. Do users perceive PUZZLE as useful?
Among favourable reactions recorded were the following:
First, PUZZLE is seen as a structuring method that formalises heuristic reasoning,
clari®es BI and helps to perform it. All participants agreed that ªit is important for
managers to understand how to perform BI based upon weak signs which is completely
different from illegal, unethical economic espionageº. Also, all agreed that interpreting
weak signs in order to transform them into intelligence is both important and necessary
before becoming actionable information.
Second, PUZZLE allows better understanding of the concept `weak signs' and orients
the environmental scanning. It shows where BI bottlenecks are in order to propose
improvements ªcurrently, a number of departments across the company collect competi-
tive information on an ad-hoc basis, and very little of it is ever turned into useful intelli-
gence. Rather than anticipating what our competitors are going to do, we found ourselves
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 146
reading photocopied snippets of what they have already doneº, commented another
participant.
Third, PUZZLE is an organisational tool to support the co-ordination of scanning, helps
selecting weak signs and interpreting activities. As a consequence, the informal activities
of scanning currently operating within companies could be reduced. ªPUZZLE avoids
overloading managers and executives by selecting only crucial pieces of informationº,
commented one participant.
Fourth, the cognitive process of individuals is heuristic. Throughout the observations, it
seems that test participants of PUZZLE follow heuristic reasoning and not algorithmic: an
idea emerges, that launches others, then other information elements are searched, and the
process is iterated during creation of puzzles, hypotheses and commentaries. As indivi-
duals scan the external environment, the new information and interpretation they acquire
in¯uence their perception of environmental changes, which in turn affects what they will
perceive in their environment and how they will scan it. That ®nding suggests that BI is a
feedback process and that interpretation activities are close to that of the targeting and
selecting of weak signs.
Fifth, Puzzle stimulates interest and helps the interpretation of weak signs based solely
upon cognition ªYou give experts a tool to go further in reaction and in re¯ection. With
PUZZLE we are concentrated more on the cognition of experts. This is a very original
work which is different from most published work or existing tools like Datamining, for
exampleº commented a participant. This suggest that interpreting weak signs is much
concerned with human activity and as with technology.
Based on the positive reactions, we can assume that test participants perceive PUZZLE
as being useful for interpreting weak signs and for improving BI.
6.4. Do users perceive puzzle as easy to use?
In addition to the previous favourable reactions, participants also perceive PUZZLE as a
helpful tool that encourages communication and discussion. They also agreed that learning
BI skills could contribute to enhancing and improving career opportunities, students'
creativity, and could help students to learn real-world skills.
Despite the positive above reaction, PUZZLE shows some weaknesses. Even though
participants recognised the value of connecting weak signs to generate working hypoth-
eses, they perceived this activity to be dif®cult. They especially perceived this activity
more dif®cult when puzzles are created individually than collectively. According to other
test participants, puzzles are not suitable when many different information elements have
to be interpreted.
In addition, it was suggested that improving the collective method could give richer
weak signs, de®ne a typology of weak signs to facilitate the puzzles creation, and
develop a library of puzzles in order to facilitate their management. Finally, other test
participants thought that PUZZLE requires creative persons for interpreting weak
signs.
Based upon the above reactions, we can infer that PUZZLE is an easy to use educational
tool and the proposed concepts seemed to be understandable. However the proposed
concepts need further improvement and re®nements (Table 1).
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 147
7. Discussion and conclusions
This paper has presented a SBIS, based on exploratory research, to support managers in
practice to better handling weak signs and to ensure success of the business strategy. This
is an elusive issue related to knowledge management that has received little academic
attention.
This paper brings three main contributions: conceptual contribution in the form of a
creative method to support interpreting weak signs, design contribution in the form of
computer tool to support this method, and empirical contribution for understanding the
process of interpreting weak signs.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 148
Table 1
Summary of ®ndings
Criteria Expected
contribution
Findings
Perceived value of
PUZZLE to BI and
interpreting weak signs
BI process ± PUZZLE clari®es the content of BI
± PUZZLE fully supports the BI process (targeting, tracking, and
interpretation of weak signs)
± BI is a feedback process
± Puzzle is method to initiate performing BI activities
± PUZZLE helps selecting of weak signs
Interpreting
weak signs
± PUZZLE is seen as a structuring method that formalises heuristic
reasoning
± Transforming weak signs into intelligence is important and
necessary before becoming actionable information
± PUZZLE allows better understanding the concept of `weak
signs' and orients environmental scanning
± The cognitive process of individuals during weak sign
interpretation appears heuristic
± PUZZLE stimulates interest and helps the interpretation of weak
signs based solely upon cognition
± Interpretation activities are close to that of targeting and
selecting of weak signs
± PUZZLE is an organisational tool to support the co-ordination of
scanning and interpreting activities
Ease of use of PUZZLE Puzzle in
operation
± PUZZLE is a helpful tool for training and for quick learning of
BI and the interpretation of weak signs
± PUZZLE contributes to improving users' and students' BI skills,
their creativity, and their career opportunities
± PUZZLE helps students to learn real-world skills
± PUZZLE facilitates communication and discussion
Weakness
of PUZZLE
± Connecting weak signs is a dif®cult individual task but less so
when collectively undertaken
± Puzzles are not suitable when many information elements have
to be interpreted.
± PUZZLE requires creative persons for interpreting weak signs
These ®ndings have implications for both research and practice. From a practical
perspective, our ®ndings encourage the use of PUZZLE by managers to perform the
process of BI. The use of a tracking form, based on `actor/theme/weak sign/ personnel
enrichment/link, transformations', continuous discussion and creation of puzzles, are all
suitable for company's use. That form constitutes a shared language for understanding BI
and that will facilitate its use in companies. PUZZLE can also be a viable training method
for managers who want to learn BI process quickly and effectively.
From a research perspective this paper supports previous researches and adds new
®ndings. This paper suggests interpreting weak signs as both a process and a
product. The process requires the transformation of seemingly unrelated information
elements into useful puzzles (product) using links in order to produce hypotheses
about potential opportunities and threats. We found that collective interpretation
produces more value than when individuals work alone. Convergence of individual
interpretations following discussions seems to be fast and easy. These ®ndings
support the previous results of El Sawy and Pauchant (1988). Moreover, this
study suggests interpreting weak signs and BI as complex processes in which indi-
viduals follow a heuristic cognitive process. The discovery of managers' dif®culties
in connecting weak signs was totally unexpected because it had never been high-
lighted before and brings new ®ndings to strategic management literature. This
®nding also contradicts most statements of previous researches (Daft and Weick,
1984; Martinsons, 1994; Attaway, 1998; Subramanian and IsHak, 1998; Freeman,
1999).
In addition, this research encourages the use of the engineering research methodology
proposed by Channel et al. (1997). This could be very helpful for researchers who want to
understand the application of SIS in an area where little underlying theory exists, or for
ill-structured problems. This methodology is very useful because of its ability to stimulate
interest, create new information, open new research directions, and to propose re®nement
of concepts.
Results of this study should be examined in light of its limitations. First, we do not
claim that this research is other than exploratory since the sample size was small, which
limits the ability to generalise the result beyond this sample. Second, we obtained data
from a single informant and that may introduce respondent bias and limit the generalisi-
bility of the results. However, as companies compete in a turbulent environment, the
results and their implications are not industry-speci®c. Third, limits of this research should
also be examined with the reliability of its results. This is concerned with the extent to
which our results are reproducible and consistent with those produced by other
approaches. This is a dif®cult question to answer in any research study, particularly one
as exploratory as ours. Our intent was to explore whether the PUZZLE approach could
provide insights into the important research question posed above, and not to engage in a
rigorous comparison of alternative BI processes and software.
Future research can take several directions. First this research suggests to continue
testing PUZZLE, through replication under different conditions and in other organisations,
in order to learn more about weak signs interpretation including: what are the frequency of
using and updating puzzles? For which enterprises the PUZZLE approach is more suita-
ble? What are the characteristics of those enterprises (size, activities, etc.)? Does the use of
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 149
the prototype result in more creative ideas and better interpretation of weak signs? Also
there is a need to apply PUZZLE to a speci®c case study such as an emerging technology,
like the web telephone.
Second, ®ndings about the dif®culty to connect weak signs to each other suggests new
research directions toward understanding the adequacy of organisational memory models
with PUZZLE. One model that is well-known and respected in this ®eld is that of Horst
Rittel's Issue Based Information Systems (IBIS) (Rittel and Kunz, 1970). The IBIS model
focuses on the articulation of key issues in the design problem. Each issue can have many
positions. A position is a statement or assertion that resolves the issue. Each of an issue's
positions, in turn, may involve one or more arguments that either support that position or
object to it. Thus each separate issue is the root of a (possibly empty) tree, with the
children of the issues being positions, and the children of the positions being arguments.
In addition, new issues raised during discussion may be posted at any time and linked into
the most appropriate nodes. Those postings serve as an organisational memory that not
only captures the ®nal decisions when trees are `closed,' but also shows the history of the
alternatives explored (Conklin and Begeman, 1988). This paper encourages future
research to apply this model with puzzle by creating a graphical web based IBIS that
allows all members of a company to read and post an issue (weak sign), positions
(enrichments and questions), and arguments (links) at any time using their web
connections. Those suggestions will help to build a complete and coherent theory around
PUZZLE for BI.
Acknowledgements
Without the support of the Centre des Etudes et de Recherches AppliqueÂes aÁ la Gestion,
Ecole SupeÂrieure des Affaires, Universite Pierre MendeÁs (France) and the Information
and Technology department at the Eindhoven Faculteit Technology Management
(The Nertherlands), this study would not have been possible. The authors gratefully
acknowledge the contributions made by these organizations, professors Omar El Sawy
and Humbert Lesca, the Editor-in-chief and anonymous reviewers.
References
Aguilar, F., 1967. Scanning the business environment. Macmillan, New York.
Ansoff, I., McDonnell, E., 1990. Implanting strategic management. 2nd ed. Prentice-Hall, Englewood Cliffs, NI.
Ansoff, H.I., 1975. Managing strategic surprise by response to weak signals. California Management Review.
XXVIII (2), 21±33.
Attaway, M.C., 1998. A Review of issues related to gathering and assessing competitive intelligence. American
Business Review 16 (1), 125±135.
Beal, R.M., 2000. Competing effectively: Environmental scanning, competitive strategy, and organisational
performance in small manufacturing ®rms. Journal of Small Business Management 38, 27±47.
Bostrom, R.P., Nagasundaram, M., 1998. Research in Creativity and GSS. Proceedings of the Thirty-First Hawaii
International Conference on System Sciences, January 6±9. vol. 6. IEEE Computer Society, Piscataway, NJ
ISBN 0-(8186)-8248-5, Best paper in minitrack, pp. 391±505
Channel, V., Lesca, H., Martinet, A.C., 1997. Vers une ingeÂnierie de la recherche en science de gestion. Revue
FrancËaise de gestion, november±december, 41±51.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 150
Choe, J-M., Lee, Y-H., Park, K.-C., 1998. The relationship model between the in¯uence factors and the strategic
applications of information systems. European Journal of Information Systems 7, 137±149.
Clarke, R., 1994. The path of development of Strategic Information Systems Theory. Available from http://
www.anu.edu.au/people/Roger.Clarke/SOS/StratISTh.html.
Conklin, J., Begeman, M., 1988. gIBIS: A Hypertext tool for Exploratory Policy Discussion. ACM Transactions
on Of®ce Information Systems 6, 4 October.
Couger, J.D., Higgins, L.F., McIntyre, S.C., 1993. (Un)structured creativity in information systems organizations.
MIS Quarterly December, 375±397.
Courtial, J.P., 1994. A coword analysis of scientometrics. Scientometrics 31, 251±260.
Cyert, R., March, J., 1963. A behavioral theory of the ®rm. Prentice-Hall, Englewood Cliffs, NJ.
Daft, R., Weick, K.E., 1984. Toward a model of organisation as interpretation systems. Academy of Management
Review 9, 284±296.
Davis, G.B., Olson, M., 1985. Management information systems: foundations, structure and development.
McGraw-Hill, New York.
Drucker, P.F., 1993. Managing in Turbulent Times. Reprint edition. Harper Collins Publishers, ISBN:
0887306160.
El Sawy, O., 1985. Personal information systems for strategic scanning in turbulent environments: can the CEO
go on-line? MIS Quarterly March, 53±60.
El Sawy, O.A., Pauchant, T., 1988. Triggers, templates and twitches in the tracking of emerging strategic issues.
Strategic Management Journal 9, 455±473.
Freeman, O., 1999. Competitor intelligence: information or intelligence. Business Information Review 16 (2),
71±77.
Futures Group, 1997. Ostriches and eagles 1997, Futures Group Inc. (August) available from http://www.tfg.com/
pubs/docs/0_Elll-97.html.
Galliers, R.D., 1991. Strategic Information Systems planning: myths reality and guidelines for successful
implementation. European Journal of Information Systems 1 (1), 55±64.
Gilad, T., Gilad, B., 1986. Business intelligence. The quiet revolution. Sloan Management Review 27, 53±62.
Groom, J.R., David, F.R., 2001. Competitive intelligence activity among small ®rms. SAM Advanced Manage-
ment Journal winter, 12±20.
Grove, A., 1999. Only the paranoid survive: How to Exploit the Crisis Points that Challenge Every Company. 1st
current edition, Bantam Books.
Hall, C., 2001. The intelligent puzzle. Competitive Intelligence Review 12 (4), 3±14.
Herring, J.P., 1998. What is intelligence analyst? Competitive intelligence Magazine 1 (2), 13±16.
Kanter, R.M., 1988. Change-Master Skills: What it takes to be creative. In: Kuhn, R.L. (Ed.). Handbook for
Creative and Innovative General Managers. McGraw-Hill, New York, NY, pp. 91±99.
Kelly, K., 1998. New rules for the new economy: ten ways the network economy is changing everything. Fourth
Estate, London.
Koneig, G., 1996. Management strategique, paradoxes, interactions et apprentissages. Nathan, Paris.
Lesca, H., Caron, M.L., 1995. Veille strateÂgique: creÂer une intelligence collective au sein de l'entrepise. Revue
FrancËaise de Gestion septembre±octobre, 58±68.
Lesca, H., 1994. Veille strategique pour le management strategique eÂtat de la question et axes de recherche.
Economies et societeÂs, seÂrie Science de Gestion 5 (20), 31±50.
Maccrimmon, K.R., Wagner, C., 1994. Stimulating ideas through creativity software. Management Science 40
(11), 1514±1532.
Martinsons, M.G., 1994. A strategic vision for managing business intelligence. Information Strategy Spring, 17±
33.
McCann, J., Gomez-Meija, L., 1992. Going on-line in the environmental scanning process. IEEE Transaction on
Engineering Management 39 (4), 394±399.
McCusker, I.C., 1992. IT effectivenessÐWhat does management need to know? The EDP Auditor Journal 3, 25±
29.
Meyer, A.D., 1991. Visual data in organizational research. Organisation Science 2 (2), 218±236.
Miller, G.A., 1956. The magical number seven, plus or minus two: some limits on our capacity for processing
information. Psychological Review 63 (2), 81±97.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 151
Moore, G., McKenna, R., 1999. Crossing the chasm: marketing and selling high-tech products to mainstream
customers. Rev ed. Perennial (Harper Collins) publisher, ISBN: 0066620023.
Nolan, J.A., 1999. It is the third millennium: do you know where your competitor is? Journal of Business Strategy
November/December, 11±15.
Norman, 1982. Learning and Memory. W.H. Freeman, San Francisco, CA.
Piaget, J., 1996. EpisteÂmologie geÂneÂtique. Press Universitaire de France.
Plsek, P.E., 1996. Working paper: models for the creative process. available from http://www.directedCreativity.
com/pages/WPModels.html.
Rittel, H., Kunz, W., 1970. Working paper. Issues as Elements of Information Systems. 131. Institute of Urban
and Regional Development, University of California, Berkeley, CA.
Rouibah, K., 1998. Puzzle: a business intelligence tool for the interpretation of anticipatory and fragmentary weak
signs. PhD Dissertation, Ecole SupeÂrieure des Affaires, Universite Pierre MendeÁs France, Grenoble, France.
Rouibah, K., Lesca, H., Malek, S., 1997. Business intelligence. A processing heuristic for weak signal. 15th
Annual International Conference of AoM/IAoM, Montreal, QueÂbec, Canada, 6±9 aouÃt 15 (1), 97±106.
Sabherwal, R., King, W., 1995. An emprical taxonomy of the decision-making process concerning strategic
applications of information systems. Journal of Management Information Systems 11 (4), 177±214.
Schneider, S.C., Meyer, A.D., 1991. Interpreting and responding to strategic issues: the impact of national culture.
Strategic Management Journal (12), 307±320.
Simon, H., 1960. The new science of management decision. Prentice-Hall, Englewood Cliffs.
Stenberg, R.J., 1988. The nature of creativity. Cambridge University Press, New York.
Subramanian, R., IsHak, S., 1998. Competititor analysis practices of US companies: an empirical investigation.
MIR Management International Review 3, 7±23.
Thomas, J.B., McDaniel, B.R., 1990. Interpreting strategic issues: effects of strategy and the information-proces-
sing structure of top management teams. Academy of Management Journal 33 (2), 286±306.
Valette, F., 1993. Le concept de Puzzle: coeur du processus d'eÂcoute prospective de l'environnement de l'en-
treprise. ThDoc ESA, France.
Wagner, J.A., Gooding, R.Z., 1997. Equivocal information and attribution: an investigation of patterns of
managerial sense making. Strategic Management Journal 18 (4), 275±286.
Walls, J.G., Widmeyer, G.R., El Sawy, O.A., 1992. Building an information system design theory for vigilant
EIS. Information System Research 3 (1), 36±59.
Wiseman, C., 1988. Strategic Information Systems. Irwin, Home-Wood, IL.
K. Rouibah, S. Ould-ali / Journal of Strategic Information Systems 11 (2002) 133±152 152
doc_367857659.pdf